1 EXECUTIVE SUMMARY. Executive Summary 1-1

Transcription

1 1 EXECUTIVE SUMMARY Distributed generation and storage technologies are playing an increasing role in the electricity system. As more distributed generation and storage systems improve their cost effectiveness and market acceptance, it will force policy makers, utility managers, and energy planners to make decisions about investing in these technologies. These investment decisions are not straight forward. There is a variety of distributed generation and storage technologies, each with different costs and benefits. In addition, the relationship between who pays and who benefits from investments in these technologies is complex. California s Self-Generation Incentive Program (SGIP) has over fourteen years of experience with installation and operation of a variety of distributed generation and storage technologies. Using cost effectiveness calculations, this report describes the costs and benefits of SGIP-eligible technologies from multiple stakeholder perspectives, including participants (Participant Cost Test), utilities (Program Administrator Cost Test), the combined perspective of participants and non-participants (Total Resource Cost Test), and society at large (Societal Total Resource Cost Test). 1 Cost effectiveness analysis can provide a clear and consistent framework for comparing the value of competing distributed generation and storage technologies against one another and against conventional grid resources. Cost effectiveness analysis also explicitly identifies the costs of resources and the associated benefits resulting from their use. The purpose of the 2015 SGIP Cost Effectiveness Study is to deliver a model that can be used in assessing the cost effectiveness of different distributed generation and storage technologies and to provide the results of a cost effectiveness evaluation of those technologies implemented using the model. The focus on this study is on SGIP technologies implemented between 2014 and This study updates and expands on a distributed generation cost effectiveness evaluation completed in January Distributed generation and storage technologies evaluated in this study can be viewed as consisting of a number of different technologies interconnected to the grid in an assortment of ways. However, distributed generation and storage technologies deployed under California s Self-Generation Incentive Program (SGIP) make up the focus of this study. For the purposes of this report, we evaluate the cost effectiveness of selected distributed generation and storage technologies eligible under the SGIP as of calendar year For this report, the Total Resource Cost Test (TRC) and the Societal Total Resource Cost (STRC) Test are calculated using similar inputs except for the discount rate applied to future benefits and costs. The TRC uses the utility s discount rate (7.5%) while the STRC uses a lower, societal discount rate (5%). 2 California Public Utilities Commission, Cost effectiveness of Distributed Generation Technologies, Itron, February 9, Executive Summary 1-1

2 Cost effectiveness of SGIP technologies are evaluated in this study following guidelines set forth in an August 2009 California Public Utilities Commission (CPUC) adopted decision on cost-benefit methodology for DG technologies COST TESTS FOR SGIP TECHNOLOGY EVALUATION SGIP technology cost effectiveness is evaluated from four perspectives: all utility customers (participants and non-participants), society, participants, and Program Administrators (PAs). 4 Based originally on the cost tests used for evaluating energy efficiency programs, these cost tests have been modified in accordance with the CPUC s 2009 adopted methodology to be applicable to distributed generation and storage technologies. The Total Resource Cost (TRC) test treats the program measures as a series of resource options. The test measures the net benefits and costs of a program and/or measure that accrue to all utility customers, both participants and non-participants. The TRC benefits are largely the avoided electric and gas costs, but include federal tax benefits or credits. The TRC costs are the costs associated with program administration, the customer measures, and increased operating costs. The TRC does not include the cost of incentives. The TRC test examines the value of the program as another way to achieve certain utility or policy goals. A positive TRC or a TRC ratio greater than 1.0 indicates that the program or measure is estimated to produce a net benefit in the utility territory over the life of the measure. The societal version of the Total Resource Cost (STRC) test looks at the overall cost effectiveness of SGIP technologies to society at large. The societal test is similar to the TRC except it uses the societal discount rate (a lower discount rate than the utility discount rate used in the TRC). 5 If the ratio of the STRC benefitsto-costs exceeds 1.0, the benefits to society exceed the costs in implementing the SGIP technology. Note that the TRC and STRC provide very similar results in this analysis. In California, both the STRC and TRC tests take into account monetary values for emissions. 6 The only difference between the STRC and TRC is the discount rate used in the tests. 7 While the model generates results for both the TRC and STRC, 3 California Public Utilities Commission, Decision Adopting Cost-Benefit Methodology for Distributed Generation, Decision , August The CPUC specifically excludes use of the Ratepayer Impact Measure (RIM) test in evaluating the cost effectiveness of DG technologies in its decision on cost-benefit methodology (see D , pg. 25) and that has been expanded to this DG and AES cost effectiveness evaluation. 5 Neither the TRC nor the STRC includes utility incentives as a benefit or a cost. From society s point of view, a rebate is a transfer from one person to another and, therefore, does not change society s benefits or costs. 6 The monetary values for emissions are derived from the E3 Avoided Cost Model dated May, The cost of carbon in this model is based on CO2 prices from the CPUC MPR Forecast. In 2014, the CO2 price is $22.50/ton, increasing to $36.97/ton in The NOx price is $6.40/lb in 2014 and $12.47/lb in The discount rate used in the STRC is 5% and in the TRC is 7.5%. Executive Summary 1-2

3 we focus primarily on STRC results. This has been done so as to remain consistent with the methodology used in the 2011 SGIP cost effectiveness report. The Participant Cost test (PCT) examines the cost effectiveness of the SGIP technology to the participant. Examples of participant benefits include electricity and gas bill savings, favorable tax treatment, or new revenue streams including utility rebates. Participant costs include increased capital outlay associated with the technology, increased operating and maintenance (O&M) costs, and fueling costs. If the benefits outweigh the costs, the technology is considered cost effective to the participant. The Program Administrator Cost (PAC) test examines the cost effectiveness of SGIP technologies from the utility perspective (noting that these costs and benefits are passed on to ratepayers). 8 It compares the net costs of participant projects (i.e., the PA costs and incentives) to other supply-side resource options available to the utility. It takes into account the costs incurred by the PAs (including incentive costs) and excludes participant costs. The PAC represents the utility s perspective (and the perspective of the utility s customers) on the net value of implementing the portfolio of projects making up the SGIP. SGIP technologies evaluated under these cost effectiveness tests include technologies eligible under the SGIP through the end of ,10 Taking into account different fuel types, this study examines the cost effectiveness of 31 different configurations of SGIP technologies including stand-alone AES systems, wind turbines, fossil-fueled as well as biogas-fueled internal combustion (IC) engines, microturbines, small-scale gas turbines and fuel cells, pressure reduction turbines (PRT), and Organic Rankine Cycle (ORC) systems. 11 Descriptions of the evaluated technologies, their operating characteristics, and costs are contained in Appendix A. Fossil-fueled technologies, other than all-electric fuel cells, are treated as having combined heat and power (CHP) capabilities. 12 Under the SGIP Handbook guidelines, technologies fueled by directed biogas or onsite biogas are not required to recover waste heat. 13 Consequently, for the cost effectiveness results presented in this report, only the natural gas-fueled technologies are modeled with CHP capabilities. 8 The study does not provide cost effectiveness results for Southern California Gas Company (SCG) even though SCG is very active in the SGIP. Due to the way in which core and non-core gas costs and prices are handled in the model, it is not possible to generate comparable gas-based cost effectiveness results for SCG. 9 The exception is solar photovoltaic (PV), which was eligible as a technology under the SGIP prior to January 1, However, in accordance with CPUC Decision , solar PV technologies were transitioned to the California Solar Initiative effective January 1, Cost effectiveness of solar PV technologies is examined in the CPUC report California Solar Initiative Cost Effectiveness Evaluation, April The 31 different configurations are shown explicitly in Table 3-1 in Section CHP refers to the production of both electric power and heat, which can be used to meet onsite electrical and thermal needs. CHP systems typically use waste heat recovery systems to capture heat generated from the power production process Self-Generation Incentive Handbook, Section 4.2.7, Minimum Operating Efficiency Requirements, pg. 44, January 1, Executive Summary 1-3

4 We have developed an updated cost effectiveness model (the 2014 SGIP Cost Effectiveness Model or 2014 SGIPce ) and use it in assessing cost effectiveness for currently eligible SGIP technologies. The model combines SGIP technology cost, performance, and financial and environmental information along with utility rate and avoided cost information. The model calculates results at the technology level as well as at utility and statewide levels. The cost-benefit results are presented in four different snapshots over time: 2014, 2017, 2020, and Individual cost and benefit components are illustrated in charts and listed in corresponding tables. The data are provided in ways to help identify the underlying causes and trends that produce specific cost effectiveness results. 1.2 KEY FINDINGS Societal Test Results We examine the STRC results of commercial-sector SGIP technologies at two years: 2014 (the current time period) and 2020 (when the SGIP is scheduled to expire). 14 Figure 1-1 illustrates a summary of the STRC results for all the evaluated commercial sector SGIP technologies at 2020 without incentives. Note that the solid horizontal line represents a benefit-cost ratio of 1.0, where an STRC of 1.0 or greater implies that the benefits exceed the costs of the SGIP technology to society. The future path of expected benefits and costs for these technologies, however, is uncertain. Given this uncertainty, and the potential noneconomic benefits associated with facilitating the market for SGIP technologies, it also makes sense to view STRC benefit-cost ratios at a lower threshold. In this instance, we have used a lower STRC threshold of The dotted horizontal line in Figure 1-1 represents the lower STRC benefit-cost ratio of 0.8. Review of the results in Figure 1-1 shows the following:» Nearly all (18 out of 26) of the evaluated SGIP technologies pass the lower STRC benefit-cost ratio of 0.8 by 2020.» SGIP technologies with an STRC benefit-cost ratio less than 0.8 in 2020 include microturbines fueled by natural gas or directed biogas, fuel cells with CHP capabilities fueled by natural gas or directed biogas, the electric-only fuel cells regardless of the fuel source; and the large storage (5 MW) technology. 14 To be consistent with the 2011 Cost Effectiveness study results, we look only at commercial sector results for key findings. Residential sector results are discussed in Section High uncertainty bounds are not unheard of in cost effectiveness analysis. For example, the 2008 California Statewide Potential Study used a TRC test of 85% to determine eligibility for program rebates. The primary focus of the Statewide Potential Study was to develop the gross and net potential estimates for electricity and gas savings in the existing and new residential, commercial, and industrial sectors. In general, the uncertainty of energy efficiency measures may be less than the uncertainty of SGIP technologies due to the shorter expected useful life of energy efficiency measures when compared to SGIP technologies. A copy of the 2008 Statewide Potential Study can be downloaded from Executive Summary 1-4

5 » Eight of the evaluated SGIP technologies had STRC benefit-cost ratios greater than 1.0. Factors that contribute to these high STRC benefit-cost ratios include no fueling costs, favorable tax treatment, and additional revenue streams (e.g., Renewable Energy Credits). Detailed results on each technology are provided in Section 6 of the report. Executive Summary 1-5

6 FIGURE 1-1: STATEWIDE SOCIETAL TOTAL RESOURCE COST (STRC) TEST RESULTS WITHOUT INCENTIVES AT 2020 Executive Summary 1-6

7 Participant Cost Test Results The 2014 PCT results are important in that they help identify SGIP technologies that may not currently be cost effective to participants and, therefore, could benefit from incentives to help overcome market risks and barriers. The 2014 SGIPce model generates the costs and benefits necessary to calculate the PCT by SGIP technology, electric investor-owned utility (IOU) service territory, market sector (e.g., residential, commercial, or government/non-profit) and geographical region (i.e., inland or coastal). As with STRC results, detailed PCT results for each evaluated SGIP technology can be found in Section 6 of the report. Figure 1-2 provides a summary snapshot of the PCT results in 2014 for commercial sector SGIP technologies without incentives. Similar to the 2020 STRC results, a solid horizontal line shows the benefitcost threshold of 1.0. However, instead of a dotted line located across the chart at a benefit-cost ratio of 0.8, the PCT dotted line is located at a benefit-to-cost ratio of 1.2. Measures with a PCT larger than 1.0 are estimated to have participant benefits greater than their participant costs. Risks and uncertainty may limit adoption of these measures. A higher PCT threshold of 1.2, however, may have sufficiently high benefitcost ratios to help overcome market barriers, risks, and uncertainty borne by participants. Key findings relative to the PCT include the following:» All but six of the SGIP technologies have 2014 PCT benefit-cost ratios less than or equal to 1.2. The technologies that had PCT benefit-cost ratios greater than 1.2 include: > Certain SGIP technologies fueled by onsite biogas, including the 200 kw microturbine, both the 2.5 MW and 7 MW gas turbines, and the 1.5 MW IC engine. > Both PRT and ORC technologies, which have no fueling costs.» SGIP technologies showing the highest PCT benefit-cost ratios are generally those without fueling costs. 16 With the exception of the electric-only fuel cells, this includes all SGIP technologies fueled by onsite biogas, wind energy systems, and PRT and ORC technologies. 16 Natural gas fueled technologies such as the 2.5 MW gas turbine and 1.5 MW IC engine also have high PCT ratios, due largely to increased bill savings tied to lower cost natural gas. Executive Summary 1-7

8 FIGURE 1-2: STATEWIDE PARTICIPANT COST TEST (PCT) RESULTS WITHOUT INCENTIVES AT 2014 Executive Summary 1-8

9 Combined STRC and PCT Results The combined STRC and PCT results provide valuable insights. The combined results identify the cross section of SGIP technologies that potentially provide high benefits to society in 2020 but that may not be cost effective to participants in Figure 1-3 shows the combined 2020 STRC and 2014 PCT results without incentives. A grey border identifies those SGIP technologies with 2020 STRC benefit-to-cost ratios greater than or equal to 0.8 and that also have 2014 PCT benefit-to-cost ratios less than or equal to 1.2. This subset of SGIP technologies represents those SGIP technologies that should ideally be targeted to receiving incentives, having both potentially high societal values in 2020 and facing market barriers in 2014 that may prevent them from achieving those high societal benefits. SGIP technologies that fall into this subset include:» The 500 kw IC engines regardless if fueled by natural gas, directed biogas, or onsite biogas; and the 1.5 MW IC engines fueled by natural gas or directed biogas.» Both the 2.5 MW and 7 MW gas turbine if fueled by natural gas, or directed biogas.» The 500 kw CHP fuel cell fueled by onsite biogas.» The 30 kw AES and the 1.5 MW wind energy technologies. Executive Summary 1-9

10 FIGURE 1-3: COMBINED STATEWIDE 2020 STRC AND 2014 PCT RESULTS WITHOUT INCENTIVES Executive Summary 1-10

11 Results on Modified Internal Rate of Return Modified internal rate of return (MIRR) represents the financial value of investments, with a higher MIRR reflecting a better investment. The 2014 SGIPce model generates MIRR values that correspond to the PCT benefit-cost results. In particular, MIRR values can be produced for each SGIP technology that reflects the financial return to the participant when there is no incentive being provided and at different levels of incentives. In general, higher incentive levels will correspond to higher financial returns to the participant. Policy makers face the question of how much incentive should be provided to different technologies. MIRR analysis helps to provide some insights into how different incentive levels may affect the financial returns to participants. The MIRR analysis allows targeting of incentive levels to help create a level playing field. For technologies with equivalent risks, the MIRR feature of the 2014 SGIPce model can be used to calculate incentive levels that provide the same MIRR for the different SGIP technologies analyzed for this study. Figure 1-4 summarizes the MIRR results that correspond to SGIP technologies without incentives and with levels of incentives expected in SGIP in We have only presented those SGIP technologies that had MIRR values greater than zero without incentives in FIGURE 1-4: COMPARISON OF MIRR FOR SGIP SYSTEMS WITHOUT INCENTIVES AND WITH 2014 INCENTIVES 17 MIRR analyses are inherently tied to existing incentive levels so as to benchmark them. Consequently, we only present MIRR results at 2014 for this summary table. Executive Summary 1-11

12 The dotted line represents the average MIRR value (approximately 10%) across the evaluated SGIP technologies when incentives are not provided. 18 The solid line represents the average MIRR value (approximately 12%) when those same SGIP technologies received incentive levels expected within SGIP in Note that technologies estimated to have a relatively high PCT without incentives (see Figure 1-2) are also shown to have a relatively high MIRR (see Figure 1-4). Policy makers can use the MIRR results to estimate incentive levels for SGIP technologies necessary to reach a given return. If a MIRR value of 12% is determined to be the target level, the SGIPce model can provide a corresponding incentive level that matches the target MIRR. 19 While the MIRR analysis helps identify possible incentive levels that match target MIRR values, ultimately the selection of incentive values is a policy decision that must take into account market risk and uncertainty that cannot be captured accurately by the model. Program Administrator Cost (PAC) Results Figure 1-5 depicts the PAC test results for SGIP systems evaluated at the statewide level for 2014 and The results are weighted by electricity sales for each of the three major electrical IOUs. 20 The 2014 results are presented using SGIP program incentives for 2014 as specified in the SGIP Handbook. The 2020 results use incentive levels forecast out to 2020 based on existing SGIP Handbook guidelines regarding annual declines in incentive levels by technology. The results presented in Figure 1-5 show that all evaluated SGIP technologies other than stand-alone energy storage have PAC benefit-cost ratios significantly higher than 1. These high PAC benefit to cost ratios result largely due to two factors occurring concurrently: high avoided electricity cost benefits being generated at the same time the technology has low or zero fueling costs. 21 As the avoided electricity costs (in the numerator) increase simultaneously with the fuel costs (in the denominator) dropping, the difference creates large benefit to cost ratios. In addition, Figure 1-5 indicates that larger sized technologies in general have higher benefit to cost ratios. This results because the larger technologies provide a disproportionality greater amount of benefit generated for each dollar of incentive paid out to the technology. The SGIP reduces the amount of incentive paid as the rebated capacity of the technology increases. However, the benefits produced and allocated to the SGIP remain proportional to the total rebated capacity. Consequently, the larger sized technologies generate a greater amount of benefits than their smaller counterparts relative to the amount 18 The 10% MIRR does not include the MIRR for technologies whose MIRR is zero or negative without an incentive. The average MIRR is calculated as an arithmetic average giving each technology equal weight. 19 The SGIPce model can generate incentive levels needed to achieve many different MIRR levels. 20 A more complete breakout of the PAC results showing individual PA results is shown in Table 6-10 in the main body of the report. 21 Note that while natural gas prices are currently low and forecast to remain low, we have also assumed that most commercial customers are not purchasing natural gas from the utility but purchasing it through gas marketers or other channels. Executive Summary 1-12

13 of incentive paid to the technology. The net result is that larger capacity SGIP technologies (which generate commensurately higher avoided cost benefits) with low or zero fuel costs and low incentive levels result in very high PAC benefit-to-cost ratios. Stand-alone storage has lower PAC benefit-to-cost ratios. In the case of the modeled 5 MW system, the 2014 and 2020 ratios are 0.81 and 1.10, respectively. For the 30 kw system, the 2014 and 2020 ratios are 0.41 and 0.71, respectively. In general, stand-alone storage shows lower PAC benefit-to-cost ratios due to the lower amount of electricity system benefits (storage is only displacing electricity over a portion of the year and the demand reductions are only pronounced in the SDG&E service territory), there are fueling costs associated with charging the storage system and the incentives paid out to storage tend to be higher than for the other SGIP technologies. FIGURE 1-5: PROGRAM ADMINISTRATOR COST RESULTS: 2014 AND 2020 (WEIGHTED) 1.3 RECOMMENDATIONS Based on the findings from this study, we make the following recommendations:» PAs and CPUC policy makers should consider using a benefit-cost ratio of 0.8 as a threshold for the STRC benefit-cost ratios and a ceiling of 1.2 for the PCT benefit-cost ratio when considering technology eligibility and incentives in the SGIP instead of a ratio of 1.0. Executive Summary 1-13

14 » The cost effectiveness results point to a class of technologies that should be considered for incentives. Incentives provided to SGIP technologies that have 2014 PCT ratios below 1.2 and 2020 STRC ratios equal to or greater than 0.8 provide value to both participants and society. SGIP technologies that meet these criteria include: > Both the 2.5 MW and 7 MW gas turbines fueled by directed biogas or natural gas. > The 500 kw IC engines regardless if fueled by natural gas, directed biogas, or onsite biogas; and the 1.5 MW IC engines fueled by natural gas or directed biogas > The commercial 30 kw stand-alone electric storage technology > The 1.5 MW wind technology.» While the model shows onsite biogas technologies as having PCT results above 1.2 without incentives, these technologies have inherent market risks that are difficult to monetize and are not taken into account in the model. Because these technologies also demonstrate high STRC ratios in 2020, they should be considered for incentive payment to help move the market for these technologies and capture the high societal relative benefits.» IC engines sized at 1,500 kw and gas turbines sized at 2,500 kw fueled by onsite biogas have 2014 PCT estimates slightly higher than 1.2 with 2020 estimates of STRC that exceed 0.8. Similarly, standalone electric storage sized at 5 MW has a 2014 PCT ratio well below 1.2 but a 2020 STRC ratio just barely below 0.8. Given the inherent uncertainty in the value of benefit and cost inputs and forecasts, these technologies should be considered for incentive payment in the SGIP.» The MIRR analysis shows that the average MIRR for SGIP technologies without incentives is 10%. In comparison, the average MIRR for SGIP technologies receiving incentives expected at 2014 levels is 12%. Consequently a MIRR target of 12% may provide a good target for setting SGIP incentive levels as this represents the average MIRR value associated with 2014 incentive levels. Ultimately, setting appropriate incentive levels is a policy decision connected to transforming the distributed generation and storage markets. Executive Summary 1-14